Realidade virtual para visualização de descargas parciais em transformadorespela técnica de emissão acústica

Abstract

In recent years, solutions for detecting, locating and visualizing partial discharges in transformers have been of great interest to power utilities. Among the solutions currently available on the market, the acoustic emission technique has been widely used in power transformers. By using this solution it is possible not only to detect, but also to locate the likely points where partial discharges occur. However, the visualization of the location of partial discharges detected by the acoustic emission technique still needs improvement. At present, partial discharges are usually depicted in figures with a low level of detail. In fact, the solutions available on the market generally represent, for example, the transformer as a hexahedron without detailing the internal components. This lack of specificity makes it more difficult for the tester to identify the actual location of the partial discharge, and also makes it more difficult to identify in which component of the transformer the partial discharge has occurred. In this context, virtual reality can be used as an aid to improve this visualization and prognosis. In other words, the combination of virtual reality with acoustic emission techniques can create a solution that provides the inspector with a better visualization of possible partial discharge points. In view of the above, this master’s dissertation proposes a methodology based on virtual reality for application to the acoustic emission technique in the process of visualizing the location of partial discharges in transformers. For comparison purposes with the proposed solution, the software AEwin™ from Physical Acoustics Corporation (PAC) was used as the current solution. According to the results, the proposed interface proves to be advantageous in comparison with the current solution, as it allows a more complete visualization of the transformer and the partial discharge points, and even makes it possible to identify the probable internal component of the transformer affected by the partial discharge. The proposed methodology thus facilitates the visualization of partial discharges in transformers. Therefore, the proposed solution presents itself as a useful and viable alternative for real applications, allowing better practices by maintenance teams.